Using current guidelines, only one-quarter of patients receiving an implantable cardiac defibrillator (ICD) for primary prevention of sudden cardiac arrest (SCA) receive appropriate ICD therapy within 5 years. PAREPET (Prediction of ARrhythmic Events with Positron Emission Tomography) identified four independent risk factors that predict SCA or ICD equivalent (SCAE) in subjects with ischemic cardiomyopathy. At optimized cut-points, the absence of these risk factors identified 38% of the cohort at very low risk of SCAE (<1%/yr). This is actually lower than the SCA rate for patients with coronary disease and mild left ventricular (LV) dysfunction (1.5-2%/yr) who are not candidates for an ICD. Thus, our goal is to prospectively determine whether these risk factors can identify a subgroup at low enough risk of SCAE to have an ICD safely withheld. PAREPET confirmed that denervated myocardium quantified with 11C-meta-hydroxyephedrine (HED) PET could predict time to SCAE. A post-hoc multivariate analysis subsequently determined that among those on optimal medical therapy, denervated myocardium, LV end-diastolic volume index (LVEDVI), and B-type natriuretic peptide (BNP) were the only independent predictors of SCAE. These parameters were independent of other PET, clinical, and demographic variables including infarct size, ejection fraction, and functional class. However, before proposing a very large clinical tria to test the potential for withholding ICD therapy among subjects predicted to be at low risk, a number of important details must be established. First, in PAREPET LVEDVI and BNP were found to be complementary to denervated myocardium based on a retrospective analysis. Thus, the independence and significance of these variables requires prospective validation. Second, HED uses a short half-life isotope that is ideal for limiting radiation exposure but requires local synthesis including a cyclotron. Clinical translation will therefore require a longer lived isotope that can be regionally produced. Finally, potentially withholding ICD therapy will require an approach for dynamic risk assessment in order to identify subsequent changes in risk. These issues will be addressed with the following Specific Aims: In subjects with ischemic cardiomyopathy on optimal medical therapy who receive an ICD for the primary prevention of SCA: Specific Aim #1 - prospectively validate whether LVEDVI and/or BNP are significant predictors of SCAE and are independent of denervated myocardium. Specific Aim #2 - determine if the 18F-labeled norepinephrine analog LMI1195 can reliably quantify denervated myocardium. Specific Aim #3 - determine whether repeat testing after a cardiac hospitalization predicts an increased risk of SCAE. This proposal will provide preliminary data for a prospective trial to test whether primary prevention ICDs can be safely withheld in subjects predicted to be at very low risk of SCAE, with cardiac hospitalizations expected provide a warning signal to reassess risk. Such a strategy would not only improve the alignment of device costs and complications with potential ICD benefit, but would almost certainly be cost-effective.